EP0104084A2 - Combinatorial weighing method and apparatus therefor - Google Patents
Combinatorial weighing method and apparatus therefor Download PDFInfo
- Publication number
- EP0104084A2 EP0104084A2 EP83305541A EP83305541A EP0104084A2 EP 0104084 A2 EP0104084 A2 EP 0104084A2 EP 83305541 A EP83305541 A EP 83305541A EP 83305541 A EP83305541 A EP 83305541A EP 0104084 A2 EP0104084 A2 EP 0104084A2
- Authority
- EP
- European Patent Office
- Prior art keywords
- weighing
- combinatorial
- articles
- participation
- machines
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G19/00—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups
- G01G19/387—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for combinatorial weighing, i.e. selecting a combination of articles whose total weight or number is closest to a desired value
- G01G19/393—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for combinatorial weighing, i.e. selecting a combination of articles whose total weight or number is closest to a desired value using two or more weighing units
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G13/00—Weighing apparatus with automatic feed or discharge for weighing-out batches of material
- G01G13/24—Weighing mechanism control arrangements for automatic feed or discharge
- G01G13/28—Weighing mechanism control arrangements for automatic feed or discharge involving variation of an electrical variable which is used to control loading or discharge of the receptacle
- G01G13/29—Weighing mechanism control arrangements for automatic feed or discharge involving variation of an electrical variable which is used to control loading or discharge of the receptacle involving digital counting
Definitions
- This invention relates to a combinatorial weighing system and, more particularly, to a combinatorial weighing system wherein a combinatorial weighing operation can continue, without suspension, even if some of the weighing stations constituting the system malfunction.
- a combinatorial weighing apparatus has a plurality of weighing machines each consisting of a weighing hopper and a weight sensor associated with the weighing hopper.
- combinatorial weighing is carried out by weighing articles which have been introduced into the weighing hoppers of the weighing machines, selecting the combination of weighing machines (referred to as the "optimum" combination") containing articles that give a total weight value equal to a target weight value or closest to the target weight value within preset allowable limits, discharging only those articles contained by the weighing hoppers of the selected weighing machines, and subsequently replenishing the emptied weighing hoppers with articles to prepare for the next weighing cycle.
- the foregoing sequence of steps is repeated to carry out a continiuous, highly accurate weighing operation automatically.
- Figs. 1 and 2 are views useful in explaining a combinatorial weighing apparatus for practicing the foregoing weighing method, in which Fig. 1 is a schematic view showing the weighing mechanism, and Fig. 2 is a block diagram of a combination control unit.
- Numeral 11 in Fig. 1 denotes a main feeder of vibratory conveyance type. Articles to be weighed are introduced into the main feeder 11 and imparted with vibratory motion for a predetermined length of time so as to be dispersed radially outward from the center of the main feeder.
- Numerals 12, 12... denote n-number of weighing stations which are arranged around the.main feeder 11 along radially extending lines to receive the articles dispersed by the main feeder.
- Each weighing station 12 includes a dispersing feeder 12a, a pool hopper 12b, a pool hopper gate 12c, a weighing hopper 12d, a weight sensor 12e, and a weighing hopper gate 12f.
- the dispersing feeder 12a comprises an independently vibratable conveyance device for feeding the articles by means of vibration, or an independently operable shutter arrangement. In either case, each dispersing feeder 12a is so arranged that an amount of articles received from the centrally located main feeder 11 can be introduced into the corresponding pool hopper 12b disposed therebelow.
- the pool hopper gate 12c is provided on each pool hopper 12b in such a manner that the articles received in the pool hopper 12b are released into the weighing hopper 12d when the pool hopper gate 12c is opened.
- Each weighing hopper 12d is provided with a weight sensor 12e of its own.
- the weight sensor 12e is operable to measure the weight of the articles introduced into the corresponding weighing hopper, and to apply an electrical signal indicative of the measured weight to the combination control unit shown in Fig. 2.
- the combination control unit selects the combination of articles (the "optimum" combination) which gives a total weight closest to the target weight value, as will be described below in further detail.
- Each weighing hopper 12d is provided with its own weighing hopper gate 12f.
- the collecting chute 13 has the shape of a funnel and is so arranged as to receive the articles from any of the circularly arrayed weighing hoppers 12d via the hopper gates 12f, which are located above the funnel substantially along its outer rim.
- the articles received by the collecting chute 13 are collected at the centrally located lower end thereof by falling under their own weight or by being forcibly shifted along the inclined wall of the funnel by a mechanical scraper or the like, which is not shown.
- the collecting chute 13 is provided with a timing hopper 14 at the lower end thereof for temporarily holding the collected articles. The arrival of an externally applied signal from a packaging machine or the like causes the timing hopper 14 to release the retained articles from the system.
- n-bit binary counter 21 has counted 2 n- 1 timing pulses TP, a total of 2 n- 1 different bit patterns, from 000...001 to 111...111, will have been generated.
- the generated bit pattern will be an indication of the abovementioned combination pattern.
- a target weight register 23, for storing a target weight value W a is connected to the arithmetic unit 26 to apply W a thereto.
- Numerals 24, 25 denote upper and lower limit setting devices, respectively, for storing preset allowable limits (namely an upper limit or maximum value Ma, and a lower limit or minimum value Mi, respectively) which are desirable for weight values.
- the minimum value Mi is set equal to the target weight value, as is customary. If it were set lower than the target weight value, the result could be delivery of articles having a total weight less thanthat intended, and complaints might ensue.
- the value X is applied to a comparator 27, whose output is connected to a proper weight counter 28.
- the comparator 27 discriminates whether the total weight value X lies in the range defined by Mi and Ma. Specifically, if the following relation holds: then the comparator 27 will increment (count up) the counter 28 by one.
- a minimum deviation register 29 for storing the minimum deviation is set automatically to the deviation A the first time only, and thereafter is updated as the conditions warrant, as will be described later. In the case where the minimum value Mi is set equal to the preset value, it is permissible to initially set the minimum deviation register 29 to the difference between the maximum value M a and the preset value.
- An optimum combination memory 30 is adapted to store the optimum combination pattern.
- Numerals 31 and 32 denote gates.
- a comparator 33 compares the deviation value A, namely the output of the arithmetic unit 26, with the prevailing minimum deviation value, denoted by B, stored in the minimum deviation register 29.
- the output of comparator 33 is such that the deviation value A is delivered for storage to the minimum deviation register 29 through the gate 31, and the content (combination pattern) of counter 21 is delivered for storage to the optimum combination memory 30.
- a drive control unit 34 which receives a signal from memory 30 indicative of the optimum combination pattern, is operable to open the weighing hopper gates 12f (Fig. 1) specified by the optimum combination pattern, so that the corresponding weighing hoppers discharge their articles into the collecting chute 13, and to open the corresponding pool hopper gates 12c so that the emptied weighing hoppers 12d may be replenished with articles.
- each of the pool hoppers 12b and weighing hoppers 12d contains a supply of the articles.
- the weight sensors 12e provided on the corresponding weighing machines measure the weights of the articles within the respective weighing hoppers and produce the weight values Wl through W10 which are sent to the combination control unit 20.
- the comparator 27 determines whether relation (3) is satisfied; if it is, then the state of the proper weight counter 28 is incremented by one.
- the content of the minimum deviation register 29 will be the minimum deviation value obtained from the 2 n- 1 combinations, and the content of the optimum combination memory 30 will be the combination pattern that gave said minimum value.
- the optimum combination is thus selected from the total of 2 n- 1 possible combination patterns.
- the drive control unit 34 opens the weighing hopper gates 12f of weighing machines corresponding to the "1" bits of the input combination pattern (namely the optimum combination pattern), whereby the articles in these weighing hoppers constituting the optimum combination are discharged into the collecting chute 13, after which the drive control unit 34 opens the corresponding pool hopper gates 12c to replenish the emptied weighing hoppers 12d with articles. Further, the dispersing feeders 12a corresponding to the emptied pool hoppers are vibrated for a fixed length of time to resupply these pool hoppers with articles.
- N_ l combinations are obtained, as already described. It is known that satisfactory accuracy can be achieved if N is a value of from seven to ten when weighing such articles as ordinary agricultural products, fishery products, industrial parts, foodstuffs and fabricated products.
- the number of combinations necessary to obtain a desirable accuracy depends upon such factors as the required accuracy, the nature of the articles and the manner in which the articles are supplied. Depending on these factors, there are cases where an automatic weighing apparatus having, say, ten weighing stations can weigh out articles to a satisfactory accuracy even when some of the weighing stations are not used. In other words, there are situations where there is no marked decline in accuracy even if the number of combinations formed is reduced to approximately half, as would be the result if N were diminished by one. This means that if a certain weighing station were to develop a malfunction, weighing could, in certain circumstances, continue without using the faulty weighing station.
- the prior-art practice is to issue an alarm and suspend the operation of the weighing machines. If the problem can be remedied without excessive delay, then the automatic weighing operation may be resumed following the necessary repairs. Where the problem is such that immediate repair is not possible, the operator manipulates a so-called "non-participation" switch to exclude the faulty weighing station from the weighing operation. Even if several of the weighing stations malfunction, therefore, the conventional method allows the weighing operation to be resumed without shutting down the weighing apparatus. Nevertheless, suspending the overall weighing operation, even temporarily, has an adverse effect of major proportions on the production line equipment preceding and following the weighing apparatus, and results in diminished weighing efficiency.
- an object of the present invention is to provide a combinatorial weighing system in which a combinatorial weighing operation need not be suspended even when a malfunction is discovered in several of the weighing stations, the weighing operation being allowed to continue using the weighing stations that remain following the exclusion of the faulty weighing stations, whereby weighing can be carried out with excellent efficiency.
- Another object of the present invention is to prevent a situation wherein a combinatorial weighing operation is allowed to continue despite inaccurate results, by automatically suspending the overall weighing operation when weighing stations, determined as being unsuited for participation in the weighing operation because of a malfunction, exceed a predetermined number.
- the foregoing objects are attained by providing a combinatorial weighing system wherein output values from weight sensors provided on respective ones of a plurality of weighing machines are monitored to determine whether each weighing machine to which a weight sensor belongs is suited for participation in combinatorial weighing. Weighing machines determined to be unsuited for participation are excluded from the combinatorial weighing operation, and combinatorial computations are performed based solely on the remaining weighing machines. When the number of weighing machines excluded exceeds a predetermined number, however, the weighing operation is halted.
- FIG. 3 A first embodiment of a combinatorial weighing apparatus according to the invention is illustrated in the block diagram of Fig. 3. We will assume that the apparatus has N weighing stations.
- the combinatorial weighing apparatus includes a malfunction detection unit 101 which, as will be described later in greater detail with reference to Fig. 4, is adapted to detect any malfunction that may develop in the weighing stations, the type of malfunction, and the weighing station in which the malfunction has occurred.
- a participation memory 102 comprises an N-bit memory for storing an N-bit signal received from the malfunction detection unit 101.
- the bit positions of the N-bit signal correspond to respective ones of the N weighing stations, and those bit positions raised to logical "1" represent the corresponding weighing stations that are capable of participating in a combination.
- the weight sensors 12e, 12e,... 12e each produce an electric signal indicative of the weight of the articles contained in the weighing hopper of the corresponding weighing station.
- These weight value signals designated Wl, W2,...
- the latter which are normally in the closed, i.e., disabled, state, correspond to the N bits of the participation memory 102 and are connected to the memory such that when the i-th bit of the participation memory 102 is logical "I", the i-th gate 103i will be enabled (opened) to deliver the weight value signal from the i-th weight sensor 12e to the multiplexer 22.
- the multiplexer 22 based on the bit pattern received from the counter (pattern generator) 21, delivers the weight values Wl, W2,... Wn successively to the arithmetic unit 26.
- the target weight value register 23 sets and stores the target weight value Wa, and applies the value to the arithmetic unit 26 where the value is subtracted from the total of the weight values Wl through Wn.
- An upper limit setting memory 104 and a lower limit setting memory 105 set and store the upper and lower limits of a desired range of values to be weighed out, the upper and lower limits values being delivered to upper and lower limit comparators 106, 107, respectively.
- the arithmetic unit 26 computes the difference A between the target weight value Wa from register 23 and the total of the weight values Wl, W2,... Wn received from the multiplexer 22. If the output A of the arithmetic unit 26 is equal to or smaller than the contents (upper limit value) of the upper limit setting memory 106, the comparator 106 delivers a logical "1" signal to an AND circuit 108. If the output A of the arithmetic unit 26 is equal to or greater than the contents (lower limit value) of the lower limit setting memory 105, the comparator 107 delivers a logical "1" signal to the AND circuit 108.
- the minimum deviation memory 29 stores the currently prevailing minimum deviation, and is adapted to read in the output of the arithmetic unit 26 when a gate 31 is enabled, whereby the state of the memory 29 is updated.
- the minimum deviation comparator 33 compares the absolute value of the output A from the arithmetic unit 26 with the contents B of the minimum deviation memory 29, enabling the gate 31 and delivering a logical "1" signal to the AND circuit 108 when the absolute value of the signal A is smaller than B.
- the AND circuit 108 whose output is connected to the input of a gate 109, takes the AND of the signals received from the upper limit comparator 106, lower limit comparator 107 and minimum deviation comparator 33.
- the gate 109 is enabled when its input from the gate circuit 108 is logical "1".
- the optimum combination memory 30 accepts the contents of the counter 21, whereby the memory is updated.
- Numeral 110 denotes an AND circuit for taking the AND between the data in the participation memory 102 and the data in the optimum combination memory 30, the output of the AND circuit 110 being connected to the drive control unit 34 and a gate 111.
- the drive control unit 34 controls the weighing machines corresponding to the "1" bits in the output of the AND circuit 110, whereby these weighing machines are opened and closed, supplied with articles and caused to discharge their articles, etc.
- the gate 111 is enabled when it receives a signal from the counter 21 indicating that all combinations have been generated, thereby delivering the output of the gate circuit 110 to an operation continuation decision unit 112.
- the operation continuation decision unit 112 decides whether it is possible to continue with a weighing operation, the decision being based on a signal from the malfunction detection unit 101 indicating the status of a malfunction, and on information representing the results of a weighing operation, received from the gate 111. If it is decided that weighing should not continue, the decision unit 112 sends a signal to an operation halt circuit 113.
- the latter is adapted to issue an alarm, such as by activating a lamp or buzzer, and to send a drive inhibit signal to the drive control unit 34.
- each weighing station can be facilitated by providing a lamp for each weighing station, each lamp indicating whether the corresponding weighing station is participating, i.e., operating normally, or not participating, i.e., malfunctioning.
- An alarm circuit 114 serves as a monitor and, in response to a signal received from the malfunction detection unit 101, produces an output indicative of the malfunctioning weighing station and of the type of malfunction.
- the malfunction detection unit 101 includes multiplexers lOla, 101e, a comparator 101b, a weight range memory 101c, a weight value selecting circuit 101d, weight abnormality memories lOlf-1, 101f-2,... 101f-n provided on corresponding ones of the weight sensors, first malfunction sensors lOlg-1, 101g-2,... lOlg-n through m-th malfunction sensors lOlh-1, 101h-2,... 101h-n provided at suitable locations in respective ones of the weighing stations, inverters 101j-1 through 101j-n, and OR gates 101m.
- the multiplexer 101a successively receives the weight values Wl, W2,... Wn produced by the weight sensors 12e, 12e,... 12e, and delivers these weight values to the comparator 101b in successive fashion.
- the weight range memory 101c stores a weight range on the basis of which a judgment can be made as to whether a weight sensor is in a normal condition. As an example, if an output from a weight sensor exceeds a zero region after a discharge command has been sent to its associated weighing hopper, then this may be taken as indicating that some of the articles have become jammed within the weighing hopper following discharge. On the other hand, if a large negative value is obtained, this can indicate that a weighing hopper has been removed, as for cleaning, without having been replaced.
- the comparator 101b compares the weight value delivered by the multiplexer 101a with the contents of the weight range memory 101c, and produces a logical "1" signal when the weight value falls outside of the stored range.
- the weight range memory 101c stores two ranges. The first is a normal weight range (referred to as the zero range) that prevails when a weighing hopper is devoid of articles. The second is a normal weight range (referred to as usage range) that prevails when articles have been introduced into the weighing hopper.
- the comparator 101b in coordination with the weighing cycle, determines whether the output value from each weight sensor 12e is within the zero range or usage range, producing the logical "1" signal when the output is outside the range.
- the weight value selecting circuit 101d supplies the multiplexer 101a with a selection signal for selecting a predetermined weight sensor.
- the multiplexer 101e which is operatively associated with the multiplexer lOla, delivers the output of the comparator 101b to the weight abnormality memories lOlf-1, 101f-2,... 101f-n in succession. Specifically, the result obtained by checking the i-th weight sensor 12e-i enters the i-th weight abnormality memory 101f-i.
- the first malfunction sensors 101g-1, 101g-2,... 101g-n through m-th malfunction sensors lOlh-1, 101h-2,...
- 101h-n are provided at prescribed positions of respective ones of the weighing stations, each of these sensors producing a "1" output when a malfunction is sensed in the corresponding weighing station.
- the output of the corresponding weight abnormality memory and the outputs from the corresponding malfunction sensors are subjected to a dot OR operation.
- the outputs from all malfunction sensors of the same type (such as the first malfunctions sensors lOlf-1 through lOlf-n) are subjected to an OR operation by a corresponding one of the OR gates 101m.
- the resulting signals relating to the respective weighing stations are delivered to an alarm circuit 114, as well as to the participation memory 102 through respective inverters 101j-1, lOlj-2,... lOlj-n. Further, the resulting signals relating to sensors of the same type are delivered to the alarm circuit 114 through the respective OR gates 101m, 101m,... 101m.
- the operation continuation decision unit 112 in Fig. 4 comprises an adder 112a, a comparator 112b, a non-participation boundary memory 112c, a tare number boundary memory 112d, a tare counter 112e, comparators 112f, 112g, and an OR gate 112h.
- the adder 112a adds the number of "1"s that result from taking the OR of the malfunction signals for each weighing station.
- the non-participation boundary memory 112c stores the number of weighing stations (from among the number N thereof) that, if no longer usable, would make it impossible to continue the weighing operation because of an inadequate number of available combinations.
- the comparator 112b compares the value in the adder 112a with the value in the non-participation boundary memory 112c and produces a "1" output when the value in the adder 112a is larger.
- the comparator 112g receives the output of the gate 111 (Fig. 3). If a combination has been formed (i.e., if all bits are not "0"), the comparator 112g clears the contents of the tare counter 112e to zero. When no combination is formed (i.e., all bits are "0"), the tare counter 112e is incremented by one step.
- the tare number boundary memory 112d stores how many consecutive times a combination must fail to be formed (i.e., how many times a combination fails to give a total weight value with the desirable limits) to make continuation of weighing impossible.
- the weighing hopper in each weighing station contains a supply of the articles, introduced from the immediately preceding stage of the apparatus, and .the weight sensors 12e, 12e,... produce the weight values Wl, W2,... Wn.
- a malfunction causes the weight value Wl to exceed the range set in the weight range memory 101c, and that a malfunction sensor in the first and third weighing stations produces a "1" output.
- the inverters lOlj-1, 101j-3 deliver outputs at logical "0", so that a bit signal expressed by 1111111010, indicating that the first and third weighing stations are unusable, is applied to the participation memory 102.
- the logic in each bit of the participation memory 102 is applied to the corresponding gates 103a, 103b,... 103n, with the result that the first and third of these gates are disabled, the others being opened.
- the first pattern produced by the counter 21 is 0000000001.
- the multiplexer 22 is adapted to accept, and deliver to the arithmetic unit 31, the signals from the gates 103a, 103b,... corresponding to the "1" bits produced by the counter 21. Since the first gate 103a is disabled by the output from the participation memory 102, as described above, the output Wl of gate 103a is, in effect, zero. This is the value accepted by the multiplexer 22 and applied to the arithmetic unit 26.
- the output A of the arithmetic unit 26 is checked by the upper and lower limit comparators 106, 107 to determine whether it falls within proper limits; if it does, then the comparators 106, 107 apply logical "1" signals to the AND circuit 108.
- the state of the counter 21 is raised to 0000000010
- the multiplexer 22 delivers the weight value W2 to the arithmetic unit 26 through the open gate 103b
- the computed value A is compared against the contents of the upper and lower limit comparators 106, 107
- the comparators apply the results of the comparison to the AND circuit 108.
- the minimum deviation comparator 33 compares the output A of the arithmetic unit 26 with the contents B of the minimum deviation register 29 and enables the gate 31 if A is smaller, whereby A is fed into the minimum deviation memory 29.
- the comparator 33 also applies a "1" signal to the AND circuit 108.
- the counter 21 provides the gate lll and the drive control unit 34 with a signal indicating that all combinations have been produced.
- the combination pattern contained in the optimum combination memory 30 will be that giving a total weight value closest to the target value within the desired weight limits.
- the drive control unit 34 takes the AND between the contents of the optimum combination memory 30 and the contents of the participation memory 102, and applies a drive command to the weighing stations on the basis of the results of the AND operation. For example, if the combination stored in the optimum combination memory 30 is 1011011101, then the output of the AND circuit 110 is 1011011000. As a result, a discharge command will be applied to the fourth, fifth, seventh, eighth and tenth weighing stations.
- comparator l12b or 112f will produce a "1" output, which is applied to an OR gate 112h.
- the OR gate 112h will deliver this signal to the operation halt circuit 113 as an operation halt command. In response, the circuit 113 halts the operation of the apparatus.
- the weighing operation is allowed to continue while automatically determining whether desirable weighing results are being obtained in a situation where the number of malfunctioning weighing stations does not exceed a set value. This makes it possible to reduce occasions in which production activity is upset by the occurrence of a malfunction.
- the malfunction sensors 101g-1,... 101g-n; ... 101h-1,... 101h-n mentioned above are of various kinds, examples of which are as follows:
- Figs. 5 and 6 illustrate a second and third embodiments of the present invention, respectively, in which portions similar to those of Fig. 3 are designated by like reference characters.
- combination patterns are generated and combinatorial processing executed with regard to all 2 n- 1 combinations irrespective of whether certain weighing stations are usable or not.
- 2 n- 1 combinations are computed with the output (weight value) of an unusable weighing station being rendered zero, the AND gate 110 taking the AND between the optimum combination pattern, at the instant the pattern is produced, and the contents of the participation memory_102.
- combinations are computed without regarding the outputs of weight sensors as zero. If a weighing station selected by a combination pattern malfunctions, the arrangement is such that the AND circuit 108 will issue a "0" output, preventing said combination from being stored in the optimum combination memory 30, even if the relation A ⁇ B should hold.
- the apparatus of the second embodiment is provided with a logic circuit 201 for computing the logical product between corresponding bits of the combination pattern prevailing in the counter 21, and the n-bit signal from the malfunction detection unit 101, and for taking the OR of the signals obtained from each logical product computation.
- the result (“1” or "0") is delivered to a NOT gate 202, the output whereof enters the AND circuit 108. Accordingly, if even one of the weighing stations selected by the "1" bits of a combination pattern is rendered useless by a malfunction, the output of the logic circuit 201 goes to logical "1", and logical "0" is applied the AND circut 108, the output whereof goes to logical "0" as a result.
- the inverters lOlj-1, 101j-2,... 101j-n shown in Fig. 4 are unnecessary in the malfunction detecting unit 101 of the embodiment illustrated in Fig. 5.
- weighing stations that malfunction are excluded from combinatorial computations. Also, rather than there being a fixed correlation between the bit positions of the counter 21 and the weighing stations, correlation is established anew whenever a malfuncting weighing station is excluded.
- correlation is established between the first, second, third, fourth, fifth, sixth, seventh and eighth bits positions of the counter 21 and the second, fourth, fifth, sixth, seventh, eighth, ninth and tenth weighing stations, respectively, with the weighing stations corresponding to the ninth and tenth bit positions being neglected.
- the apparatus of the third embodiment is provided with a correlation generator 301 which, using information pertaining to malfunctioning weighing stations, generates the correlation between the bits of counter 21 and the weighing stations.
- a combinatorial weighing operation need not be suspended even when a malfunction is discovered in several of the weighing stations, the weighing operation being allowed to continue using the weighing stations that remain following the exclusion of the faulty weighing stations, whereby weighing can be carried out in an extremely efficient manner.
- the invention also prevents a situation wherein a combinatorial weighing operation is allowed to continue despite inaccurate results. This is accomplished by automatically suspending the overall weighing operation when weighing stations, determined as being unsuited for participation in the weighing operation because of a malfunction, exceed a predetermined number, or when all combinations of the weighing stations give total weight values that fall outside predetermined allowable limits consecutively n times.
- each weight value produced by the weight sensors is monitored, and when a weight value fails to fall within a predetermined range, the weighing station to which the corresponding weight sensor belongs is judged to be defective, the weighing being then being automatically excluded from combinatorial computations. Therefore, if a weighing hopper should be removed for cleaning, or if a cleaned weighing hopper should happen not to be reattached to the weighing apparatus, the associated weight sensor will not introduce an excessively negative value into the computations to upset the weighing operation. Maintenance may be facilitated by providing a lamp or the like for each weighing station, the lamp indicating whether the weighing station is operating normally or not.
- a weighing station is judged to be defective and is then automatically excluded from combinatorial computations whenever the corresponding weight sensor produces a weight value that is outside of the predetermined limits. Therefore, if for some reason the operator should wish to remove a weighing station from partication, means that cause the weight value to fall outside the predetermined limits may be provided. Such means can be a weight or metal fitting that forces the weight value out of said limits. The same results can be obtained by disconnecting the proper wire lead or the like. A weighing station so excluded from participation in a weighing operation may then be inspected or subjected to maintenance with facility.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Weight Measurement For Supplying Or Discharging Of Specified Amounts Of Material (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
- This invention relates to a combinatorial weighing system and, more particularly, to a combinatorial weighing system wherein a combinatorial weighing operation can continue, without suspension, even if some of the weighing stations constituting the system malfunction.
- A combinatorial weighing apparatus has a plurality of weighing machines each consisting of a weighing hopper and a weight sensor associated with the weighing hopper. According to a known combinatorial weighing method, combinatorial weighing is carried out by weighing articles which have been introduced into the weighing hoppers of the weighing machines, selecting the combination of weighing machines (referred to as the "optimum" combination") containing articles that give a total weight value equal to a target weight value or closest to the target weight value within preset allowable limits, discharging only those articles contained by the weighing hoppers of the selected weighing machines, and subsequently replenishing the emptied weighing hoppers with articles to prepare for the next weighing cycle. The foregoing sequence of steps is repeated to carry out a continiuous, highly accurate weighing operation automatically.
- Figs. 1 and 2 are views useful in explaining a combinatorial weighing apparatus for practicing the foregoing weighing method, in which Fig. 1 is a schematic view showing the weighing mechanism, and Fig. 2 is a block diagram of a combination control unit. Numeral 11 in Fig. 1 denotes a main feeder of vibratory conveyance type. Articles to be weighed are introduced into the main feeder 11 and imparted with vibratory motion for a predetermined length of time so as to be dispersed radially outward from the center of the main feeder. Numerals 12, 12... denote n-number of weighing stations which are arranged around the.main feeder 11 along radially extending lines to receive the articles dispersed by the main feeder. Each
weighing station 12 includes a dispersingfeeder 12a, apool hopper 12b, apool hopper gate 12c, a weighinghopper 12d, aweight sensor 12e, and a weighinghopper gate 12f. The dispersingfeeder 12a comprises an independently vibratable conveyance device for feeding the articles by means of vibration, or an independently operable shutter arrangement. In either case, each dispersingfeeder 12a is so arranged that an amount of articles received from the centrally located main feeder 11 can be introduced into thecorresponding pool hopper 12b disposed therebelow. Thepool hopper gate 12c is provided on eachpool hopper 12b in such a manner that the articles received in thepool hopper 12b are released into the weighinghopper 12d when thepool hopper gate 12c is opened. Each weighinghopper 12d is provided with aweight sensor 12e of its own. Theweight sensor 12e is operable to measure the weight of the articles introduced into the corresponding weighing hopper, and to apply an electrical signal indicative of the measured weight to the combination control unit shown in Fig. 2. The combination control unit then selects the combination of articles (the "optimum" combination) which gives a total weight closest to the target weight value, as will be described below in further detail. Each weighinghopper 12d is provided with its own weighinghopper gate 12f. A drive control unit, shown in Fig. 2, upon receiving the signals from each of the weight sensors, produces a signal to open only the weighinghopper gates 12f of those weighinghoppers 12d that give the optimum combination, thesegates 12f discharging the articles from the corresponding weighinghoppers 12d into acommon chute 13 where they are collected together. Thecollecting chute 13 has the shape of a funnel and is so arranged as to receive the articles from any of the circularly arrayed weighinghoppers 12d via thehopper gates 12f, which are located above the funnel substantially along its outer rim. The articles received by thecollecting chute 13 are collected at the centrally located lower end thereof by falling under their own weight or by being forcibly shifted along the inclined wall of the funnel by a mechanical scraper or the like, which is not shown. Thecollecting chute 13 is provided with atiming hopper 14 at the lower end thereof for temporarily holding the collected articles. The arrival of an externally applied signal from a packaging machine or the like causes thetiming hopper 14 to release the retained articles from the system. - Reference will now be had to the block diagram of Fig. 2 for a description of the combination control unit. Numeral 20 denotes the combination control unit which includes an n-bit (n=10)
counter 21 for counting timing pulses TP of a predetermined frequency, and for generating all combinations of n-number of the weighing hoppers. These combinations will also be referred to as "combination patterns" where appropriate. Specifically, for n-number of weighing hoppers, n combinations are possible when each combination is composed of one weighing hopper from the total of n weighing hoppers, n(n-l)/2! combinations are possible when each combination is composed of two weighing hoppers selected from said total, and, in general, n(n-1)(n-2)...(n-r+1)/r! combinations are possible when each combination is composed of r-number of weighing hoppers selected from said total of n weighing hoppers. Accordingly, when the n-bitbinary counter 21 has counted 2n-1 timing pulses TP, a total of 2n-1 different bit patterns, from 000...001 to 111...111, will have been generated. Therefore, if correlation is established between the first bit and the first weighing hopper, between the second bit and the second weighing hopper, and between third through n-th bits and the third through n-th weighing hoppers, then the generated bit pattern will be an indication of the abovementioned combination pattern. - A
multiplexer 22, in accordance with the output bit pattern of thecounter 21, provides anarithmetic unit 26 with read values (indicative of the weight of the article batches) from theweight sensors 12e of predetermined weighing hoppers. For instance, if the value of the count (the bit pattern) incounter 21 is 1000101011 when n=10, then thearithmetic unit 26 will receive the weight value outputs Wl, W2, W3, W4, W6, W10 from theweight sensors 12e attached to the first, second, fourth, sixth and tenth weighing hoppers, respectively. Atarget weight register 23, for storing a target weight value Wa, is connected to thearithmetic unit 26 to apply Wa thereto.Numerals - The
arithmetic unit 26 computes, and delivers a signal indicative of, the total weight ΣWi (=X) of the weight values received from themultiplexer 22, and also computes the difference between the total weight Σ Wi and the preset value Wa. Thearithmetic unit 26 produces a signal A indicating the absolute value of the computed difference. More specifically, thearithmetic unit 26 performs the operations:comparator 27, whose output is connected to aproper weight counter 28. Thecomparator 27 discriminates whether the total weight value X lies in the range defined by Mi and Ma. Specifically, if the following relation holds:comparator 27 will increment (count up) thecounter 28 by one. Aminimum deviation register 29 for storing the minimum deviation is set automatically to the deviation A the first time only, and thereafter is updated as the conditions warrant, as will be described later. In the case where the minimum value Mi is set equal to the preset value, it is permissible to initially set theminimum deviation register 29 to the difference between the maximum value Ma and the preset value. Anoptimum combination memory 30 is adapted to store the optimum combination pattern.Numerals comparator 33 compares the deviation value A, namely the output of thearithmetic unit 26, with the prevailing minimum deviation value, denoted by B, stored in theminimum deviation register 29. When the inequality A<B holds, the output ofcomparator 33 is such that the deviation value A is delivered for storage to theminimum deviation register 29 through thegate 31, and the content (combination pattern) ofcounter 21 is delivered for storage to theoptimum combination memory 30. - When the state of
counter 28 is one or more, adrive control unit 34, which receives a signal frommemory 30 indicative of the optimum combination pattern, is operable to open the weighinghopper gates 12f (Fig. 1) specified by the optimum combination pattern, so that the corresponding weighing hoppers discharge their articles into thecollecting chute 13, and to open the correspondingpool hopper gates 12c so that the emptied weighinghoppers 12d may be replenished with articles. - The operation of the weighing apparatus will now be described. At the beginning, each of the
pool hoppers 12b and weighinghoppers 12d contains a supply of the articles. Theweight sensors 12e provided on the corresponding weighing machines measure the weights of the articles within the respective weighing hoppers and produce the weight values Wl through W10 which are sent to thecombination control unit 20. The n-bit (n=10)counter 21 counts the timing pulses TP having the predetermined frequency to produce 2n -1 combination patterns. Thus, when the first timing pulse TP arrives and is counted, the content ofcounter 21 becomes 0000000001. As a result, themultiplexer 22 sends the first weight value signal W1, from theweight sensor 12e provided on the first weighing hopper, to thearithmetic unit 26, which responds by performing the operation specified by Eqs. (1) and (2) above, thereby producing the total weight value Σ Wi of the combination, as well as the deviation A (=|W1-Wa|) between the total weight value Σ Wi and the target weight value Wa. Since thegates minimum deviation register 29, and the content (the combination pattern 0000000001) of n-bit counter 21 is stored in theoptimum combination memory 30.Comparator 27 compares the total weight Σ Wi (=X) against the maximum value Ma and the minimum value Mi, and increments theproper counter 28 when the relation Mi < X < M holds. Thenceforth, when the second timing pulse TP is generated, the pulse is counted bycounter 21, whose state (combination pattern) is incremented to 0000000010. Consequently, the weight value output W2 of theweight sensor 12e provided on the second weighing hopper is delivered to thearithmetic unit 26 which then performs the operations of equations (1) and (2) to produce the signals indicative of the total weight Z wi (=X) and of the deviation value A (A = |W2-Wa|). Thecomparator 27 then determines whether relation (3) is satisfied; if it is, then the state of theproper weight counter 28 is incremented by one. Thecomparator 33, meanwhile, compares the deviation value A with the state B (=|W1-Wa|) of theminimum deviation register 29. If the relation A>B holds, then neither theregister 29 nor theoptimum combination memory 30 is updated; if A<B holds, the deviation value A is transferred to and stored inregister 29, and the state ofcounter 21 is transferred to and stored in theoptimum combination memory 30. The operation described above is repeated until all 2n-1 combinations have been generated. At such time the content of theminimum deviation register 29 will be the minimum deviation value obtained from the 2n-1 combinations, and the content of theoptimum combination memory 30 will be the combination pattern that gave said minimum value. The optimum combination is thus selected from the total of 2n-1 possible combination patterns. - If the value of the count in
counter 28 is one or more, thedrive control unit 34 opens the weighinghopper gates 12f of weighing machines corresponding to the "1" bits of the input combination pattern (namely the optimum combination pattern), whereby the articles in these weighing hoppers constituting the optimum combination are discharged into thecollecting chute 13, after which thedrive control unit 34 opens the correspondingpool hopper gates 12c to replenish the emptied weighinghoppers 12d with articles. Further, the dispersingfeeders 12a corresponding to the emptied pool hoppers are vibrated for a fixed length of time to resupply these pool hoppers with articles. - This completes one combinatorial weighing cycle, which may be repeated as often as required, to provide batches of the articles, each batch having a total weight equal or closest to the preset value. It should be noted that when the state of the
proper weight counter 28 is zero in the foregoing operation, articles are not discharged and each of the weighing machines must be supplemented with articles to resume the combinatorial computations. - With the above-described combinatorial weighing method, even articles which exhibit a great difference in weight from one to another can be weighed out to within an error that is less than the article unit weight, thereby making it possible to achieve a highly accurate weighing operation.
- In an automatic weighing apparatus of the aforementioned combinatorial type in which combinations are formed among N weight values, 2N_l combinations are obtained, as already described. It is known that satisfactory accuracy can be achieved if N is a value of from seven to ten when weighing such articles as ordinary agricultural products, fishery products, industrial parts, foodstuffs and fabricated products.
- The number of combinations necessary to obtain a desirable accuracy depends upon such factors as the required accuracy, the nature of the articles and the manner in which the articles are supplied. Depending on these factors, there are cases where an automatic weighing apparatus having, say, ten weighing stations can weigh out articles to a satisfactory accuracy even when some of the weighing stations are not used. In other words, there are situations where there is no marked decline in accuracy even if the number of combinations formed is reduced to approximately half, as would be the result if N were diminished by one. This means that if a certain weighing station were to develop a malfunction, weighing could, in certain circumstances, continue without using the faulty weighing station.
- When a specific weighing station malfunctions, the prior-art practice is to issue an alarm and suspend the operation of the weighing machines. If the problem can be remedied without excessive delay, then the automatic weighing operation may be resumed following the necessary repairs. Where the problem is such that immediate repair is not possible, the operator manipulates a so-called "non-participation" switch to exclude the faulty weighing station from the weighing operation. Even if several of the weighing stations malfunction, therefore, the conventional method allows the weighing operation to be resumed without shutting down the weighing apparatus. Nevertheless, suspending the overall weighing operation, even temporarily, has an adverse effect of major proportions on the production line equipment preceding and following the weighing apparatus, and results in diminished weighing efficiency.
- Accordingly, it has been contemplated to merely present some indication of weighing stations detected as having malfunctioned, while allowing weighing to proceed without any suspension in operation. However, a problem that arises with this technique is that the combinatorial weighing computations do not provide satisfactory results when the malfunctioning weighing stations grow large in number. In other words, an excessive number of combinations provide total weight values that fall outside the allowable limits, the end result being a weighing operation having poor accuracy.
- Accordingly, an object of the present invention is to provide a combinatorial weighing system in which a combinatorial weighing operation need not be suspended even when a malfunction is discovered in several of the weighing stations, the weighing operation being allowed to continue using the weighing stations that remain following the exclusion of the faulty weighing stations, whereby weighing can be carried out with excellent efficiency.
- Another object of the present invention is to prevent a situation wherein a combinatorial weighing operation is allowed to continue despite inaccurate results, by automatically suspending the overall weighing operation when weighing stations, determined as being unsuited for participation in the weighing operation because of a malfunction, exceed a predetermined number.
- According to the present invention, the foregoing objects are attained by providing a combinatorial weighing system wherein output values from weight sensors provided on respective ones of a plurality of weighing machines are monitored to determine whether each weighing machine to which a weight sensor belongs is suited for participation in combinatorial weighing. Weighing machines determined to be unsuited for participation are excluded from the combinatorial weighing operation, and combinatorial computations are performed based solely on the remaining weighing machines. When the number of weighing machines excluded exceeds a predetermined number, however, the weighing operation is halted.
- Other features and advantages of the present invention will be apparent from the following description taken in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the figures thereof.
-
- Fig. 1 is a schematic view showing the weighing mechanism of a combinatorial weighing apparatus according to the prior art;
- Fig.-2 is a block diagram of a combination control unit in the weighing apparatus of the prior art;
- Fig. 3 is a block diagram illustrating an embodiment of a combinatorial weighing apparatus according to the present invention;
- Fig. 4 is a block diagram illustrating a malfunction detection unit and operation continuation decision unit in the apparatus of Fig. 3; and
- Figs. 5 and 6 are block diagrams illustrating second and third embodiments, respectively, of a combinatorial weighing apparatus according to the present invention.
- Reference will now be had to the accompanying drawings to describe embodiments of the present invention.
- A first embodiment of a combinatorial weighing apparatus according to the invention is illustrated in the block diagram of Fig. 3. We will assume that the apparatus has N weighing stations.
- The combinatorial weighing apparatus includes a
malfunction detection unit 101 which, as will be described later in greater detail with reference to Fig. 4, is adapted to detect any malfunction that may develop in the weighing stations, the type of malfunction, and the weighing station in which the malfunction has occurred. Aparticipation memory 102 comprises an N-bit memory for storing an N-bit signal received from themalfunction detection unit 101. The bit positions of the N-bit signal correspond to respective ones of the N weighing stations, and those bit positions raised to logical "1" represent the corresponding weighing stations that are capable of participating in a combination. Theweight sensors respective gates participation memory 102 and are connected to the memory such that when the i-th bit of theparticipation memory 102 is logical "I", the i-th gate 103i will be enabled (opened) to deliver the weight value signal from the i-th weight sensor 12e to themultiplexer 22. - The
multiplexer 22, based on the bit pattern received from the counter (pattern generator) 21, delivers the weight values Wl, W2,... Wn successively to thearithmetic unit 26. - The target
weight value register 23 sets and stores the target weight value Wa, and applies the value to thearithmetic unit 26 where the value is subtracted from the total of the weight values Wl through Wn. An upperlimit setting memory 104 and a lowerlimit setting memory 105 set and store the upper and lower limits of a desired range of values to be weighed out, the upper and lower limits values being delivered to upper andlower limit comparators - The
arithmetic unit 26 computes the difference A between the target weight value Wa fromregister 23 and the total of the weight values Wl, W2,... Wn received from themultiplexer 22. If the output A of thearithmetic unit 26 is equal to or smaller than the contents (upper limit value) of the upperlimit setting memory 106, thecomparator 106 delivers a logical "1" signal to an ANDcircuit 108. If the output A of thearithmetic unit 26 is equal to or greater than the contents (lower limit value) of the lowerlimit setting memory 105, thecomparator 107 delivers a logical "1" signal to the ANDcircuit 108. Theminimum deviation memory 29 stores the currently prevailing minimum deviation, and is adapted to read in the output of thearithmetic unit 26 when agate 31 is enabled, whereby the state of thememory 29 is updated. Theminimum deviation comparator 33 compares the absolute value of the output A from thearithmetic unit 26 with the contents B of theminimum deviation memory 29, enabling thegate 31 and delivering a logical "1" signal to the ANDcircuit 108 when the absolute value of the signal A is smaller than B. - The AND
circuit 108, whose output is connected to the input of agate 109, takes the AND of the signals received from theupper limit comparator 106,lower limit comparator 107 andminimum deviation comparator 33. Thegate 109 is enabled when its input from thegate circuit 108 is logical "1". Thecounter 21, serving as a combination pattern generator, generates combination patterns each composed of N bits. If N=10, the counter will generate 2 -1 patterns successively, from 0000000001 to 1111111111. The i-th bit of each pattern corresponds to the i-th weighing machine. For example, if thecounter 21 produces a bit pattern 1001001101, where N=10, then themultiplexer 22 will provide thearithmetic unit 26 with weight values W1, W3, W4, W7 and W10. When thegate 109 is enabled, theoptimum combination memory 30 accepts the contents of thecounter 21, whereby the memory is updated. -
Numeral 110 denotes an AND circuit for taking the AND between the data in theparticipation memory 102 and the data in theoptimum combination memory 30, the output of the ANDcircuit 110 being connected to thedrive control unit 34 and a gate 111. Thedrive control unit 34 controls the weighing machines corresponding to the "1" bits in the output of the ANDcircuit 110, whereby these weighing machines are opened and closed, supplied with articles and caused to discharge their articles, etc. The gate 111 is enabled when it receives a signal from thecounter 21 indicating that all combinations have been generated, thereby delivering the output of thegate circuit 110 to an operationcontinuation decision unit 112. - The operation
continuation decision unit 112, which will be described in detail below, decides whether it is possible to continue with a weighing operation, the decision being based on a signal from themalfunction detection unit 101 indicating the status of a malfunction, and on information representing the results of a weighing operation, received from the gate 111. If it is decided that weighing should not continue, thedecision unit 112 sends a signal to anoperation halt circuit 113. The latter is adapted to issue an alarm, such as by activating a lamp or buzzer, and to send a drive inhibit signal to thedrive control unit 34. It should be noted that maintenance can be facilitated by providing a lamp for each weighing station, each lamp indicating whether the corresponding weighing station is participating, i.e., operating normally, or not participating, i.e., malfunctioning. Analarm circuit 114 serves as a monitor and, in response to a signal received from themalfunction detection unit 101, produces an output indicative of the malfunctioning weighing station and of the type of malfunction. - Reference will now be had to Fig. 4 to describe the details of the
malfunction detection unit 101 and operationcontinuation decision unit 112. Themalfunction detection unit 101 includes multiplexers lOla, 101e, a comparator 101b, aweight range memory 101c, a weight value selecting circuit 101d, weight abnormality memories lOlf-1, 101f-2,... 101f-n provided on corresponding ones of the weight sensors, first malfunction sensors lOlg-1, 101g-2,... lOlg-n through m-th malfunction sensors lOlh-1, 101h-2,... 101h-n provided at suitable locations in respective ones of the weighing stations,inverters 101j-1 through 101j-n, and ORgates 101m. The multiplexer 101a successively receives the weight values Wl, W2,... Wn produced by theweight sensors weight range memory 101c stores a weight range on the basis of which a judgment can be made as to whether a weight sensor is in a normal condition. As an example, if an output from a weight sensor exceeds a zero region after a discharge command has been sent to its associated weighing hopper, then this may be taken as indicating that some of the articles have become jammed within the weighing hopper following discharge. On the other hand, if a large negative value is obtained, this can indicate that a weighing hopper has been removed, as for cleaning, without having been replaced. The comparator 101b compares the weight value delivered by the multiplexer 101a with the contents of theweight range memory 101c, and produces a logical "1" signal when the weight value falls outside of the stored range. It should be noted that theweight range memory 101c stores two ranges. The first is a normal weight range (referred to as the zero range) that prevails when a weighing hopper is devoid of articles. The second is a normal weight range (referred to as usage range) that prevails when articles have been introduced into the weighing hopper. The comparator 101b, in coordination with the weighing cycle, determines whether the output value from eachweight sensor 12e is within the zero range or usage range, producing the logical "1" signal when the output is outside the range. The weight value selecting circuit 101d supplies the multiplexer 101a with a selection signal for selecting a predetermined weight sensor. The multiplexer 101e, which is operatively associated with the multiplexer lOla, delivers the output of the comparator 101b to the weight abnormality memories lOlf-1, 101f-2,... 101f-n in succession. Specifically, the result obtained by checking the i-th weight sensor 12e-i enters the i-thweight abnormality memory 101f-i. Thefirst malfunction sensors 101g-1, 101g-2,... 101g-n through m-th malfunction sensors lOlh-1, 101h-2,... 101h-n are provided at prescribed positions of respective ones of the weighing stations, each of these sensors producing a "1" output when a malfunction is sensed in the corresponding weighing station. For each weighing station, the output of the corresponding weight abnormality memory and the outputs from the corresponding malfunction sensors are subjected to a dot OR operation. In addition, the outputs from all malfunction sensors of the same type (such as the first malfunctions sensors lOlf-1 through lOlf-n) are subjected to an OR operation by a corresponding one of theOR gates 101m. The resulting signals relating to the respective weighing stations are delivered to analarm circuit 114, as well as to theparticipation memory 102 throughrespective inverters 101j-1, lOlj-2,... lOlj-n. Further, the resulting signals relating to sensors of the same type are delivered to thealarm circuit 114 through the respective ORgates - The operation
continuation decision unit 112 in Fig. 4 comprises an adder 112a, a comparator 112b, anon-participation boundary memory 112c, a tarenumber boundary memory 112d, atare counter 112e,comparators non-participation boundary memory 112c stores the number of weighing stations (from among the number N thereof) that, if no longer usable, would make it impossible to continue the weighing operation because of an inadequate number of available combinations. The comparator 112b compares the value in the adder 112a with the value in thenon-participation boundary memory 112c and produces a "1" output when the value in the adder 112a is larger. Thecomparator 112g receives the output of the gate 111 (Fig. 3). If a combination has been formed (i.e., if all bits are not "0"), thecomparator 112g clears the contents of thetare counter 112e to zero. When no combination is formed (i.e., all bits are "0"), thetare counter 112e is incremented by one step. The tarenumber boundary memory 112d stores how many consecutive times a combination must fail to be formed (i.e., how many times a combination fails to give a total weight value with the desirable limits) to make continuation of weighing impossible. - The operation of the invention will now be described with reference to Figs. 3 and 4.
- The weighing hopper in each weighing station contains a supply of the articles, introduced from the immediately preceding stage of the apparatus, and .the
weight sensors weight range memory 101c, and that a malfunction sensor in the first and third weighing stations produces a "1" output. Under these conditions, the inverters lOlj-1, 101j-3 deliver outputs at logical "0", so that a bit signal expressed by 1111111010, indicating that the first and third weighing stations are unusable, is applied to theparticipation memory 102. The logic in each bit of theparticipation memory 102 is applied to thecorresponding gates - The first pattern produced by the
counter 21 is 0000000001. Themultiplexer 22 is adapted to accept, and deliver to thearithmetic unit 31, the signals from thegates counter 21. Since thefirst gate 103a is disabled by the output from theparticipation memory 102, as described above, the output Wl ofgate 103a is, in effect, zero. This is the value accepted by themultiplexer 22 and applied to thearithmetic unit 26. Thegate 31 initially is open, so that the difference between Wl (= 0) and the target weight value, computed by thearithmetic unit 26, is delivered to theminimum deviation memory 29. At this time, the output A of thearithmetic unit 26 is checked by the upper andlower limit comparators comparators circuit 108. - With the generation of the next timing pulse TP, the state of the
counter 21 is raised to 0000000010, themultiplexer 22 delivers the weight value W2 to thearithmetic unit 26 through theopen gate 103b, the computed value A is compared against the contents of the upper andlower limit comparators circuit 108. Theminimum deviation comparator 33 compares the output A of thearithmetic unit 26 with the contents B of theminimum deviation register 29 and enables thegate 31 if A is smaller, whereby A is fed into theminimum deviation memory 29. Thecomparator 33 also applies a "1" signal to the ANDcircuit 108. Thereafter, the state ofcounter 21 is incremented by the generation of the next timing pulse TP, 0000000011 is produced as the output combination pattern, and processing proceeds just as described above. Eventually, thecounter 21 will produce a 1111111111 combination pattern. When the related processing has been performed, one combinatorial computation cycle ends. When an output A from thearithmetic unit 26 falls within the set limits and, moreover, A<B is found to hold during the processing for the combinatorial computation cycle, the output of the ANDcircuit 108 goes to logical "1" to enable thegate 109. When this occurs, the state of thecounter 21 is fed intooptimum combination memory 30. When the final pattern 1111111111 is generated by thecounter 21 and the related processing is completed, thecounter 21 provides the gate lll and thedrive control unit 34 with a signal indicating that all combinations have been produced. At such time, the combination pattern contained in theoptimum combination memory 30 will be that giving a total weight value closest to the target value within the desired weight limits. Thedrive control unit 34 takes the AND between the contents of theoptimum combination memory 30 and the contents of theparticipation memory 102, and applies a drive command to the weighing stations on the basis of the results of the AND operation. For example, if the combination stored in theoptimum combination memory 30 is 1011011101, then the output of the ANDcircuit 110 is 1011011000. As a result, a discharge command will be applied to the fourth, fifth, seventh, eighth and tenth weighing stations. - In the event that a large number of the weighing stations malfunction so that a combination giving a desirable combined weight value does not exist owing to an inadequate number of available combinations, the output of the AND
circuit 108 will never rise to logical "1" during the entire combinatorial computation process. In consequence, when thecounter 21 produces the signal indicating that all combinations have been generated, all bits in theoptimum combination memory 30 will be "0". As a result, the tare counter l12e in the operationcontinuation decision unit 112 is incremented by one count. If the number of weighing stations rendered useless by a malfunction exceeds a set value, or if the count in thetare counter 112e exceeds a set value, then comparator l12b or 112f will produce a "1" output, which is applied to an OR gate 112h. The OR gate 112h will deliver this signal to theoperation halt circuit 113 as an operation halt command. In response, thecircuit 113 halts the operation of the apparatus. - Therefore, according to the present invention, if some of the N weighing stations of the weighing apparatus should become unusable because of a malfunction, the weighing operation is allowed to continue while automatically determining whether desirable weighing results are being obtained in a situation where the number of malfunctioning weighing stations does not exceed a set value. This makes it possible to reduce occasions in which production activity is upset by the occurrence of a malfunction.
- The
malfunction sensors 101g-1,... 101g-n; ... 101h-1,... 101h-n mentioned above are of various kinds, examples of which are as follows: - (a) A sensor for detecting whether electric current is flowing into an electromagnet during drive. This sensor is used in a case where the dispersing
feeder 12a (Fig. 1) is of the feeding conveyance type. - (b) A limit switch or photoelectric tube provided at the ends of the stroke of a shutter for detecting whether the shutter is open or closed. This sensor is used when the dispersing feeder is of the shutter-type construction.
- (c) A limit switch provided at the closing position of the
pool hopper gate 12c ofpool hopper 12b, and at the closing position of the weighinghopper gate 12f of weighinghopper 12d, so as to be closed when the corresponding gate is closed. This makes it possible to detect whether articles have become jammed between the hopper and gate. - (d) In a case where a.weighing hopper is opened and closed by one revolution of a cam, a radially extending slit is formed in a slit cam that rotates in unison with the first-mentioned cam. To confirm that the slit has returned to a home position, a sensor is provided for detecting a beam of light projected through the slit by a photoelectric tube. Another form of sensor is provided at the stroke end of a link lever for opening and closing a gate, the sensor detecting any abnormality in the gate opening and closing operation. Other sensors include one for measuring drive motor current to detect any overload, a temperature sensor for sensing an abnormal temperature internally of the weighing apparatus, etc.
- Figs. 5 and 6 illustrate a second and third embodiments of the present invention, respectively, in which portions similar to those of Fig. 3 are designated by like reference characters.
- In the embodiment of Fig. 3, combination patterns are generated and combinatorial processing executed with regard to all 2n-1 combinations irrespective of whether certain weighing stations are usable or not. In other words, 2n-1 combinations are computed with the output (weight value) of an unusable weighing station being rendered zero, the AND
gate 110 taking the AND between the optimum combination pattern, at the instant the pattern is produced, and the contents of the participation memory_102. In the embodiment illustrated in Fig. 5, however, combinations are computed without regarding the outputs of weight sensors as zero. If a weighing station selected by a combination pattern malfunctions, the arrangement is such that the ANDcircuit 108 will issue a "0" output, preventing said combination from being stored in theoptimum combination memory 30, even if the relation A<B should hold. More specifically, the apparatus of the second embodiment is provided with alogic circuit 201 for computing the logical product between corresponding bits of the combination pattern prevailing in thecounter 21, and the n-bit signal from themalfunction detection unit 101, and for taking the OR of the signals obtained from each logical product computation. The result ("1" or "0") is delivered to aNOT gate 202, the output whereof enters the ANDcircuit 108. Accordingly, if even one of the weighing stations selected by the "1" bits of a combination pattern is rendered useless by a malfunction, the output of thelogic circuit 201 goes to logical "1", and logical "0" is applied the ANDcircut 108, the output whereof goes to logical "0" as a result. It should be noted that the inverters lOlj-1, 101j-2,... 101j-n shown in Fig. 4 are unnecessary in themalfunction detecting unit 101 of the embodiment illustrated in Fig. 5. - In the third embodiment of the present invention, shown in Fig. 6, weighing stations that malfunction are excluded from combinatorial computations. Also, rather than there being a fixed correlation between the bit positions of the
counter 21 and the weighing stations, correlation is established anew whenever a malfuncting weighing station is excluded. By way of example, if the first and third weighing stations malfunction, then correlation is established between the first, second, third, fourth, fifth, sixth, seventh and eighth bits positions of thecounter 21 and the second, fourth, fifth, sixth, seventh, eighth, ninth and tenth weighing stations, respectively, with the weighing stations corresponding to the ninth and tenth bit positions being neglected. As a result, whereas 210- 1 combination patterns would need to be generated ordinarily, it will suffice if only 28_1 combination patterns are produced in the arrangement of Fig. 6. To achieve the foregoing, the apparatus of the third embodiment is provided with acorrelation generator 301 which, using information pertaining to malfunctioning weighing stations, generates the correlation between the bits ofcounter 21 and the weighing stations. - In accordance with the present invention as described and illustrated hereinabove, a combinatorial weighing operation need not be suspended even when a malfunction is discovered in several of the weighing stations, the weighing operation being allowed to continue using the weighing stations that remain following the exclusion of the faulty weighing stations, whereby weighing can be carried out in an extremely efficient manner. The invention also prevents a situation wherein a combinatorial weighing operation is allowed to continue despite inaccurate results. This is accomplished by automatically suspending the overall weighing operation when weighing stations, determined as being unsuited for participation in the weighing operation because of a malfunction, exceed a predetermined number, or when all combinations of the weighing stations give total weight values that fall outside predetermined allowable limits consecutively n times.
- Further, according to the present invention, each weight value produced by the weight sensors is monitored, and when a weight value fails to fall within a predetermined range, the weighing station to which the corresponding weight sensor belongs is judged to be defective, the weighing being then being automatically excluded from combinatorial computations. Therefore, if a weighing hopper should be removed for cleaning, or if a cleaned weighing hopper should happen not to be reattached to the weighing apparatus, the associated weight sensor will not introduce an excessively negative value into the computations to upset the weighing operation. Maintenance may be facilitated by providing a lamp or the like for each weighing station, the lamp indicating whether the weighing station is operating normally or not.
- As described above, a weighing station is judged to be defective and is then automatically excluded from combinatorial computations whenever the corresponding weight sensor produces a weight value that is outside of the predetermined limits. Therefore, if for some reason the operator should wish to remove a weighing station from partication, means that cause the weight value to fall outside the predetermined limits may be provided. Such means can be a weight or metal fitting that forces the weight value out of said limits. The same results can be obtained by disconnecting the proper wire lead or the like. A weighing station so excluded from participation in a weighing operation may then be inspected or subjected to maintenance with facility.
- As many apparently widely different embodiments of the present invention can be made without departing from the spirit and scope thereof, it is to be understood that the invention is not limited to the specific embodiments thereof except as defined in the appended claims.
Claims (11)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP163698/82 | 1982-09-20 | ||
JP57163698A JPS5952719A (en) | 1982-09-20 | 1982-09-20 | Combinational metering system |
Publications (3)
Publication Number | Publication Date |
---|---|
EP0104084A2 true EP0104084A2 (en) | 1984-03-28 |
EP0104084A3 EP0104084A3 (en) | 1985-09-11 |
EP0104084B1 EP0104084B1 (en) | 1989-06-14 |
Family
ID=15778909
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP83305541A Expired EP0104084B1 (en) | 1982-09-20 | 1983-09-20 | Combinatorial weighing method and apparatus therefor |
Country Status (5)
Country | Link |
---|---|
US (1) | US4609058A (en) |
EP (1) | EP0104084B1 (en) |
JP (1) | JPS5952719A (en) |
AU (1) | AU549660B2 (en) |
DE (1) | DE3380071D1 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0243064A2 (en) * | 1986-04-17 | 1987-10-28 | Ishida Scales Mfg. Co. Ltd. | Combinational weighing system |
EP0255777A2 (en) * | 1986-07-28 | 1988-02-10 | Package Machinery Company | Metal detector testing process and apparatus |
EP1016856A2 (en) * | 1998-12-28 | 2000-07-05 | Ishida Co., Ltd. | Combinatorial weighing system |
EP3617671B1 (en) | 2018-08-31 | 2022-03-30 | ISHIDA CO., Ltd. | Combination weighing device |
Families Citing this family (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0663813B2 (en) * | 1984-06-14 | 1994-08-22 | 株式会社石田衡器製作所 | Weighing device |
JPS63231228A (en) * | 1987-03-19 | 1988-09-27 | Ishida Scales Mfg Co Ltd | Quantitative combinational weighing machine |
JPH0610628B2 (en) * | 1988-12-20 | 1994-02-09 | 株式会社寺岡精工 | Manual combination weigher |
US4874049A (en) * | 1989-02-03 | 1989-10-17 | Kee Equipment And Engineering | Automatic weighing method and apparatus |
US5148841A (en) * | 1989-06-27 | 1992-09-22 | Serac Group | Filling apparatus for filling receptacle with metered weights |
US4976326A (en) * | 1989-12-22 | 1990-12-11 | R. J. Reynolds Tobacco Company | Weighing system |
US5411607A (en) * | 1993-11-10 | 1995-05-02 | Novamax Technologies Holdings, Inc. | Process and composition for sealing anodized aluminum surfaces |
US5753866A (en) * | 1994-05-13 | 1998-05-19 | Ishida Co., Ltd. | Combinational weigher |
JPH0953976A (en) * | 1995-08-09 | 1997-02-25 | Ishida Co Ltd | Combination method counter |
JPH08210908A (en) * | 1995-10-30 | 1996-08-20 | Ishida Co Ltd | Combination measuring device |
US20050060958A1 (en) * | 2003-08-29 | 2005-03-24 | Delta And Pine Land Company | Seed handling systems and methods |
JP4714514B2 (en) * | 2005-06-30 | 2011-06-29 | 大和製衡株式会社 | Combination weighing device |
JP5279577B2 (en) * | 2009-03-30 | 2013-09-04 | 大和製衡株式会社 | Combination weigher and weighing packaging system including the same |
DE102014204625A1 (en) * | 2014-03-13 | 2015-09-17 | Robert Bosch Gmbh | Method and device for determining a carbon dioxide content of an ambient air |
ES2611057B1 (en) * | 2015-11-02 | 2018-02-08 | Girnet Internacional, S.L. | Container for transporting and unloading products, especially fruit and vegetable products, and combination weighing machine comprising said containers |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2067861A (en) | 1979-12-24 | 1981-07-30 | Yamato Scale Co Ltd | Weighing device |
GB2077067A (en) | 1980-05-30 | 1981-12-09 | Yamato Scale Co Ltd | Weighing apparatus |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3004617A (en) * | 1958-06-30 | 1961-10-17 | Baldwin Lima Hamilton Corp | Electrically controlled batching system |
US4202466A (en) * | 1973-07-19 | 1980-05-13 | Henry Simon Limited | Method of and apparatus for dispensing predetermined proportions of two or more materials |
US4310893A (en) * | 1979-12-12 | 1982-01-12 | Reliance Electric Company | Digital scale |
JPS5746127A (en) * | 1980-09-03 | 1982-03-16 | Ishida Scales Mfg Co Ltd | Processing method for poor combination in combined weighing or combined counting |
JPS57151828A (en) * | 1981-03-14 | 1982-09-20 | Ishida Scales Mfg Co Ltd | Automatic zero point adjustment prohibitor |
JPS57171216A (en) * | 1981-04-15 | 1982-10-21 | Yamato Scale Co Ltd | Combination balance |
JPS58158523A (en) * | 1982-03-15 | 1983-09-20 | Yamato Scale Co Ltd | Combined scale |
-
1982
- 1982-09-20 JP JP57163698A patent/JPS5952719A/en active Granted
-
1983
- 1983-09-20 DE DE8383305541T patent/DE3380071D1/en not_active Expired
- 1983-09-20 AU AU19267/83A patent/AU549660B2/en not_active Ceased
- 1983-09-20 EP EP83305541A patent/EP0104084B1/en not_active Expired
- 1983-09-20 US US06/533,975 patent/US4609058A/en not_active Expired - Lifetime
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2067861A (en) | 1979-12-24 | 1981-07-30 | Yamato Scale Co Ltd | Weighing device |
GB2077067A (en) | 1980-05-30 | 1981-12-09 | Yamato Scale Co Ltd | Weighing apparatus |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0243064A2 (en) * | 1986-04-17 | 1987-10-28 | Ishida Scales Mfg. Co. Ltd. | Combinational weighing system |
EP0243064A3 (en) * | 1986-04-17 | 1988-07-27 | Ishida Scales Mfg. Co. Ltd. | Combinational weighing system |
EP0255777A2 (en) * | 1986-07-28 | 1988-02-10 | Package Machinery Company | Metal detector testing process and apparatus |
EP0255777A3 (en) * | 1986-07-28 | 1989-04-26 | Package Machinery Company | Metal detector testing process and apparatus |
EP1016856A2 (en) * | 1998-12-28 | 2000-07-05 | Ishida Co., Ltd. | Combinatorial weighing system |
US6373001B1 (en) | 1998-12-28 | 2002-04-16 | Ishida Co., Ltd. | Weighing, packaging and inspecting system |
EP1016856A3 (en) * | 1998-12-28 | 2003-05-14 | Ishida Co., Ltd. | Combinatorial weighing system |
EP3617671B1 (en) | 2018-08-31 | 2022-03-30 | ISHIDA CO., Ltd. | Combination weighing device |
Also Published As
Publication number | Publication date |
---|---|
US4609058A (en) | 1986-09-02 |
EP0104084B1 (en) | 1989-06-14 |
DE3380071D1 (en) | 1989-07-20 |
JPS5952719A (en) | 1984-03-27 |
EP0104084A3 (en) | 1985-09-11 |
AU549660B2 (en) | 1986-02-06 |
AU1926783A (en) | 1984-03-29 |
JPH0554051B2 (en) | 1993-08-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4609058A (en) | Combinatorial weighing method and apparatus therefor | |
US4442911A (en) | Method and apparatus for controlling zero adjustment of weight sensor | |
EP0063056B1 (en) | Combinatorial weighing method and apparatus therefor | |
EP0084968B1 (en) | Weighing method and apparatus therefor | |
EP0071473B1 (en) | Combinatorial batching of articles | |
US4494619A (en) | Combinatorial weighing apparatus with bulk and fine feed | |
US4467880A (en) | Combinatorial weighing apparatus | |
US4396078A (en) | Weighing or counting method with some unfilled hoppers | |
US4529049A (en) | Combinatorial weighing or counting method | |
US4465149A (en) | Method of zero adjustment for combinatorial weighing or counting system | |
EP0064554B1 (en) | Combination weighing method and apparatus | |
EP0127326A2 (en) | Combinatorial weighing method and apparatus | |
EP0074773B1 (en) | Combinatorial weighing method | |
EP0103476A2 (en) | Combinatorial weighing method and apparatus therefor | |
EP0119854B1 (en) | Combinatorial weighing method and apparatus | |
US4574897A (en) | Combinatorial weighing apparatus | |
US4625817A (en) | Method of zero adjustment for combinatorial weighing or counting system | |
EP0100223A2 (en) | Combinatorial weighing methods and apparatus | |
US4527646A (en) | Combination weighing machine | |
JPH0251131B2 (en) | ||
JPS59198323A (en) | Combination weighing method | |
EP0097529B1 (en) | Combinatorial weighing methods and apparatus | |
EP0093607B1 (en) | Combinatorial measuring method and apparatus | |
US4516643A (en) | Combinatorial weighing method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
AK | Designated contracting states |
Designated state(s): DE FR GB IT |
|
PUAL | Search report despatched |
Free format text: ORIGINAL CODE: 0009013 |
|
AK | Designated contracting states |
Designated state(s): DE FR GB IT |
|
17P | Request for examination filed |
Effective date: 19860204 |
|
17Q | First examination report despatched |
Effective date: 19870129 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): DE FR GB IT |
|
REF | Corresponds to: |
Ref document number: 3380071 Country of ref document: DE Date of ref document: 19890720 |
|
ET | Fr: translation filed | ||
ITF | It: translation for a ep patent filed |
Owner name: ING. ZINI MARANESI & C. S.R.L. |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed | ||
ITTA | It: last paid annual fee | ||
REG | Reference to a national code |
Ref country code: GB Ref legal event code: IF02 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 20020910 Year of fee payment: 20 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: GB Payment date: 20020918 Year of fee payment: 20 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 20020925 Year of fee payment: 20 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GB Free format text: LAPSE BECAUSE OF EXPIRATION OF PROTECTION Effective date: 20030919 |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: PE20 |